CPC G06Q 30/0613 (2013.01) [G06F 16/583 (2019.01); G06Q 30/0627 (2013.01); G06T 7/97 (2017.01); G06V 10/255 (2022.01); G06V 20/20 (2022.01); G06V 20/41 (2022.01); H04N 23/45 (2023.01); G06F 3/0482 (2013.01)] | 18 Claims |
1. An object-recognizing retail purchase system, comprising:
an automatically adjustable camera rig comprising a plurality of movable cameras, wherein the plurality of movable cameras are automatically moved by a camera control platform according to characteristics of an object within a view field of the plurality of movable cameras;
a first input for receiving images from the automatically adjustable camera rig;
a second input for receiving a plurality of scraped network images regarding a plurality of purchasable objects, wherein the scraped network images include new images for newly released products by using automated validation scripts;
a first computing memory for storing an object profile for each of the plurality of purchasable objects, wherein each of the object profiles comprises at least data from the first input regarding the object within the field of view and data from the second input;
a purchasing platform at least partially present on a mobile device and comprising at least one computer processor having resident thereon non-transitory computing code which, when executed by the at least one computing processor, causes to be performed the steps of:
receiving an image of a viewed object within a view field of a mobile device camera of the mobile device, wherein the image of viewed object is assigned one or more vectors forming at least height and width aspect ratios;
gray-scaling the image of the viewed object;
comparing the gray-scaled image to ones of the object profiles until a matched product is obtained, wherein the comparing includes determining the matched product by matching stored aspect ratios to the formed at least height and width aspect ratios that are formed from the assigned one or more vectors, and the matching includes comparing a new width ratio value to the stored width ratio values and returning the closest match when the same object at a different distance produces the same result, and wherein the data to the first input comprises gray-scaling to create haar-like features and patterns detected in pixels of the received image to determine attributes associated with the viewed object that are used in the comparison until the matched product is obtained, and; and
providing a purchase link via a graphic user interface in the mobile device suitable to enable a purchase of the matched product from at least one third party.
|